How does initiative work in BG, IWD and PT?
Alonso
Member Posts: 806
The manuals say initiative is used, but they say nothing about how it actually works. I guess it must be related to the AD&D 2ed rules, but those rules allow for many possibilities.
Post edited by Alonso on
0
Comments
in 2nd edition they use weapon speeds for initiative, so if something has a weapon speed of 5 or a casting time of 5, it will take 5/10ths of a round before you start swinging/spell goes off
now with initiative in the neverwinter nights series, im not 100% sure if its implemented properly, in 3rd edition dnd, whoever has the higher initiative goes first, and everyone starts out flat footed until it their turn, which i want to say does exist in NWN but im not 100% sure, other than that, initiative as far as i know doesnt do much else since NWN games are in real time
The 2nd edition PnP manual includes many other factors that affect initiative. The most important one is the initiative roll, each creature makes an initiative roll to determine initiative. Has that been left out altogether in the PC games? Also, what about being hasted, slowed, etc.? Nothing of that affects initiative?
As for slow and haste, I'm pretty sure slow effects just double all the frames, and haste effects cut them in half; they divide and multiply rather than add and subtract.
Tagging @Grond0, @kjeron, and @Grammarsalad to see if they have any insight into initiatives and speed factor.
if you have a weapon speed of 1 or 0, you are going to start attacking pretty quick after you start the attack action, if you have weapon speed of 10, its going to take a while before your first hit starts landing after the attack option
as apposed to multiple attacks, i do not think you roll initiative separately for each attack, i would almost bet money its only off the first attack
but as semiticgod said, with haste lowering frames, and slow doubling them, that could also affect it as well, apparently haste says it gives you a +2 bonus to the roll
if you really want to test it out, play test someone using a dagger, and see how fast they actually start getting attack rolls after you give them an attack command, and then after that, give that character a two handed sword or halberd and do the same thing, and you will see a huge difference in how fast your attack rolls actually start rolling after you do the attack action
1) Initiative only applies to weapon attacks in BG. Other actions, like drinking a potion or casting a spell, always take place at the expected time in the round. Thus a spell with casting time 5 will be started immediately and completed halfway through the round (after 5 of the 10 segments).
2) For weapon attacks, initiative either adds or subtracts from the weapon speed. Thus you could have a speed of 0 (and I don't think it's possible to go below that), but you won't always attack immediately when you attempt an attack (even if that is in a new combat round). Conversely you could have a slightly slower weapon, but still attack immediately as a result of a good initiative roll.
3) It's difficult by simple observation to determine exactly how great the initiative effect is, but my best guess is that it gives a range of +/- 3. That means that if you have a fast weapon (like a dagger) and someone else has a slow weapon (like a halberd) you can guarantee you will attack first (assuming you're careful to ensure both you and the enemy are in a fresh attack round). Most of the time though the gap in weapon speeds is not that great and you can't guarantee attacking first (but can still get huge advantages in combat by utilizing a small weapon speed advantage).
4) I doubt that subsequent attacks in a round are affected by initiative, although I can't totally rule that out.
5) I'm not sure if slow and haste have any effect on initiative, but I suspect not (other than indirectly by increasing or compressing the length of time actions take).
6) It's not directly relevant to initiative, but I'll mention that projectile weapons all have a fast weapon speed in the game (probably 1) and not their published values. You'll therefore note that you normally shoot pretty early in the round after initiating combat.
7) There are other things that can override the initiative roll. If, for instance, you have 10 APR the game has to squeeze all those into a 10 segment round. I don't think attacks are made more often than 1 per segment, so that means that even a slow weapon speed will attack immediately in the round.
8) The attack animations will also distort the combined results of weapon speed and initiative. I'm not sure exactly what the causal relationship is, but the game always tries to make actual attacks at the same time as an animation. However, you never make more than one fake attack when beginning combat in a new attack round before making an actual attack (if you interrupt an attack and then restart it without waiting for a new attack round you can make multiple fake attacks). That means that even the slowest weapons (like a 2-handed sword) will never attack slower than about the middle of the round. Conversely the fastest weapons will always attack on the first animation and therefore won't ever be affected by the full range of the potential initiative penalty.
@Grond0's knowledge is way, way higher than mine, but I will say that after running countless of BG1 runs and almost always starting with ranged weaponry, crossbows do have a slower first attack on average than bows. Kiting with a xbow with only proficient pip usually means that about half of the times (subjective) make you wait a while before you see the attack animation, and quite often Imoen with her shortbow fires first.
Even Imoen with her shortbow ends up waiting a second or so into her round before firing every now and then, but that's more rare than for my Charname (who is very often a xbow wielding dwarf).
I haven't noticed any problems with kiting using crossbows myself. The only missile weapon that I've found significantly harder to use when kiting is a throwing axe. That is considerably slower and makes it much more difficult to reliably get enough distance to attack something and retreat before they reach you. I am aware that fact rather undercuts my argument about the difference for projectile weapons being that they don't have fake attacks (the same applies to other missile weapons as well), but hey we all need some unsolved mysteries to ponder over
I may be completely wrong though since sometimes personal bias moves into my experience and "calculations" when comparing charname to other NPCs. "Gawd, why isn't he firing - The damn Ogre is getting closer!! Quick, KITE BACK - KITE BACK!"
BG/IWD/PT: This applies to the first round, but we don't know if it also affects the following rounds.
BG/IWD/PT: The games are based on the "Individual initiative" method.
BG/IWD/PT: Initiative doesn't just determine the order in which the creatures act, it also determines the precise moment in which they act. But we don't know how this works.
BG/IWD/PT: Works the same.
BG/IWD/PT: The only modifiers that we know that are used are casting time and speed factor. Slow and haste are not used. About the others, we don't know. We don't know how initiative works for monsters that attack with natural weapons.
BG/IWD/PT: Unknown.
BG/IWD/PT: Unknown.
BG/IWD/PT: Unknown.
BG/IWD/PT: Unknown.
@subtledoctor as I explained earlier, the stated weapon speeds for projectile weapons are not reflected in the game mechanics. That's extremely easy to demonstrate. Just equip a heavy crossbow on one character and fire at something - you'll see the shot is fired very quickly despite the supposed weapon speed of 10.
It's also pretty simple to demonstrate that initiative does indeed exist in relation to weapons (unlike spells), even if calculating the exact mechanic is not simple at all. The easiest way to see that is to equip a melee weapon with a mid-range weapon speed - for instance a long sword with speed of 5 (equip it on a character without weapon speed bonuses). Attack someone else in your party and see how long it is from initiating the action until the real attack. You will quickly notice that in some cases the real attack takes place fairly early in the round in the first attack animation. In other cases the real attack takes place later on in the second attack animation. I explained earlier that very slow weapons like a halberd will also attack on the second animation - but in that case they will never attack on the first, unlike the long sword. Conversely, very fast weapons will always attack on the first animation and never the second. If you observe closely though you will see that the first animation itself does not take place at exactly the same time every round - sometimes it is immediate, but sometimes after a pause (which again is initiative in action, though it's not quite as easy to see the difference as between the first and second animation).
Because of the complexities introduced by multiple APRs, if you are testing with a character with more than one attack per round, I suggest you only attack once and then withdraw briefly so that when you attack again it's on a new round.
I would recommend against pursuing this detailed comparison with PnP though - the way initiative is treated in BG is fundamentally different from PnP and making a detailed comparison may be more confusing than enlightening.
"I have compared the PnP rules with what we know about the PC games so far. The goal is to clarify as many "unknowns" as possible:
PnP rules The initiative roll determines who acts first in any given combat round.
BG/IWD/PT: This applies to the first round, but we don't know if it also affects the following rounds.
We do know - it affects every round. The uncertainty expressed before was whether initiative applied to multiple attacks after the first in a round. While that's difficult to test I'm fairly sure that it doesn't.
PnP rules Initiative is normally determined with a single roll for each side in a conflict. There are also two optional methods that can be used to determine initiative. Each of these optional methods breaks the group action down into more individual initiatives. These two optional methods are "Group initiative" and "Individual initiative".
BG/IWD/PT: The games are based on the "Individual initiative" method.
BG does use individual initiative, but what is meant by that is very different from the PnP methodology.
PnP rules To determine the initiative order for a round of combat, roll 1d10 for each creature in the battle. The lowest roll wins the initiative. The remaining creatures act in ascending order of initiative.
BG/IWD/PT: Initiative doesn't just determine the order in which the creatures act, it also determines the precise moment in which they act. But we don't know how this works.
I think we do - intiative acts as a modifier to the weapon speed of melee weapons to determine when in the round they strike. That's a fundamentally different method to PnP where initiative determines the order of attacks between combatants within a shared round. In BG initiative modifies just your own attacks with no reference to what the opponent is doing - an opponent's personal rounds can be out of sync with yours, so it would not be possible for a single initiative roll to apply to both.
PnP rules If two creatures roll the same number for initiative, everything happens simultaneously —all attack rolls, damage, spells, and other actions are completed before any results are applied. It is possible for a wizard to be slain by goblins who collapse from his sleep spell at the end of the round.
BG/IWD/PT: Works the same.
See above comment. BG uses a fundamentally different system - initiative in BG does not mean the same thing at all as in PnP (in the former initiative is a modifier to when you personally act; in the latter initiative is a determinant of the order in which both you and your opponent act).
PnP rules MODIFIERS TO INITIATIVE:
Hasted: –2
Slowed: +2
On higher ground: –1
Set to receive a charge: –2
Wading or slippery footing: +2
Wading in deep water: +4
Foreign environment: +6
Hindered (tangled, climbing, held): +3
Attacking with weapon: Weapon speed
Breath weapon: +1
Casting a spell: Casting time
Creature size (Monsters attacking with natural weapons only)
Tiny: 0
Small: +3
Medium: +3
Large: +6
Huge: +9
Gargantuan: +12
Innate spell ability: +3
Magical Items
Miscellaneous Magic: +3
Potion: +4
Ring: +3
Rod: +1
Scroll: Casting time of spell
Stave: +2
Wand: +3
BG/IWD/PT: The only modifiers that we know that are used are casting time and speed factor. Slow and haste are not used. About the others, we don't know. We don't know how initiative works for monsters that attack with natural weapons.
The type of weapon does not affect initiative (though it will affect weapon speed). I think it's misleading to refer to speed factor as modifying initiative - it seems clear to me that it's the other way round in BG. Initiative does not modify casting time in BG.
PnP rules When multiple attacks are the result of different attack forms—claws and a bite or bite and tail or a ranger with his two-weapon combat ability for example—the attacks all occur at the same time. The creature resolves all of its attacks in initiative order.
BG/IWD/PT: Unknown.
BG is different to PnP - each attack is individually determined, rather than all being done at once.
PnP rules When the attacks are true multiples —using the same weapon more than once— as in the case of a highly skilled fighter, the attacks are staggered. Everyone involved in the combat completes one action before the second (or subsequent) attack roll is made.
BG/IWD/PT: Unknown.
Same - each attack is individually determined.
PnP rules If fighters on both sides in a battle were able to attack twice in the round, their first attacks would occur according to the initiative roll. Their second attacks would come after all other attacks, and would then alternate according to the initiative roll.
BG/IWD/PT: Unknown.
Different system. In BG your attacks are determined purely according to your own individual round, without regard to what your opponent is doing in their round (though there can be indirect effects from the opponent, e.g. whenever you're hit you are staggered slightly which results in losing attacks yourself if you are constantly being hit).
PnP rules Magical Weapon Speeds: Each bonus point conferred by a magical weapon reduces the speed factor of that weapon by 1. A sword +3 reduces the weapon speed factor by 3, for example. When a weapon has two bonuses, the lesser one is used. No weapon can have a speed factor of less than 0.
BG/IWD/PT: Unknown.
In principle the same system is used. There have been some inconsistencies in the past about which of two bonuses is used, but those are minor.
@subtledoctor I don't need to make large assumptions - I'm happy to rely on evidence
I appreciate that the way people play the game is very different. I only play no-reload and normally solo. I also never use healing items and make pretty limited use of buffs. That means that I pay far more attention to when weapon attacks take place than most people. However, for the purposes of testing, anyone can check when actual attacks take place in a round - the easiest way to do that is to turn on the battle text in the game.
I agree it can be difficult to tell when a new round starts - which is why I gave you a methodology to make that clear (have one character in your party attack another; withdraw between attacks and wait briefly to ensure your next attack is at the start of a new round). Also do turn on false attacks for the purpose of testing, if you've disabled that. It doesn't change the timing, but does make it clearer visually there is a variation in the speed of attacks - avoiding any need for stopwatches.
I previously set out why the variation in attack speed appears to be designed and not a random programming glitch (slow melee weapon - never attacks early in the round; fast melee weapon - always attacks early in the round; medium melee weapon - sometimes attacks early, sometimes not).
I don't know if you've read my other posts in the thread, but I've said that the comparison with PnP is not helpful. What initiative means in PnP is fundamentally different from in BG and Bioware might have been better choosing a different word for it - but they didn't. Reading from my BG2 original manual, this is what was said about initiative: "In Baldur's Gate II initiative is used as a random variation on how quickly characters can initiate their attacks or spells. It adjusts the speed factor of a spell or weapon slightly."
Occam's razor tells me a rather different story to you. If the developers specifically say they did something then the chances are high that they have done it. That's not always the case of course, I can think of a number of instances where what is said in the manual is incorrect. In this instance that does seem to be the case with initiative for spells - either the plan to randomly vary the length of spells was deliberately abandoned or the coding for that was not correctly completed. However, in the case of weapons, what the manual says aligns with actual game behavior.
I've explained how you can easily verify that there is a random component to the speed of weapon attacks and clarified that this is intended behavior by the original developers and not just some mistake in the programming. I believe the evidence for the existence of a random initiative factor is compelling, but if you still think there is some other way to explain the actual game behavior then I will of course consider that.
I accept that it's not easy to trace rounds in the game - largely as a result of rounds being applied separately to every individual. However, the method I've suggested does ensure the next attack is at the start of a new round. You can do the same test in an alternative way though by turning on autopause for end of round - you might find using that makes it easier to see what's happening. The game only keeps track of combat rounds when someone is actually in combat. For that reason testing this behavior is more easily done by attacking someone in your party (they don't go hostile and hence don't ever have their own combat rounds to complicate things). Even with autopause active you'll note that before you attack the game never pauses - outside of combat or taking another action that triggers a combat round (like casting a spell, activating a scroll or drinking a potion) the game does not keep track of rounds. Once you initiate an attack though the round starts and the autopause will kick in to tell you when the round has finished.
I'm used to estimating for myself when rounds start and finish (both for the PC and opponents), as that's necessary to leverage weapon speed advantages in combats. As an illustration I might have a fighter armed with a dagger attack a gnoll armed with a halberd. I know that the random initiative variation is not sufficient to overcome the weapon speed differential in that combat, so I will always strike first in the round and can then withdraw before coming back to attack in the next round to get another free hit. However, if I get the timing wrong and come back in too quickly, the combat round for the gnoll won't yet have ended and I can arrive just as his actual attack occurs. In a 1vs1 situation I would very rarely make that mistake, but things do get more complicated in groups. It's not easy to know exactly how close an enemy has to be to activate a combat round, so if you're kiting a group round it's quite possible to try and run past an enemy with a slow weapon only to find that their combat round started when you were bypassing another creature a few seconds earlier and they're able to get a real attack in on you as you run by.
If you look closely you will see that the text feedback does closely correlate with animation behavior (such as weapon swinging, staggering and blood splashes) and there is clearly intended to be a link between combat results and the animation. I would agree that the link with the animation doesn't always work perfectly in practice though, particularly where there are high APRs. However, painful experience over many years (particularly seeing the game over screen thousands of times
I agree that initiative is a slightly misleading word for it as it relates to a rather different behavior than PnP initiative. However, I wouldn't describe it as utterly stupid. Above, I referred to the way I can reliably defeat enemies with much slower weapon speeds without ever being hit myself. If there were no random variation I could reliably defeat the vast majority of melee enemies in the game in the same way. Weapon speed is already a far more important factor in the game than is commonly recognized (discussions on which weapon is best almost never refer to this, but it has a significant effect on the balance of damage dealt and received). Without the random initiative factor though it would become even more important - sufficiently so I think to unduly affect the overall balance of the game.
If you wanted to explore modding options related to initiative I think it would be interesting to actually enable the original intention of the developers and apply initiative to spells as well as weapons. One of the reasons why spell casters are generally reckoned to be strong characters is the reliability with which you can cast. Thus, you might be in a position in the existing game where you can reliably finish off an almost dead opponent with a magic missile before it can attack you (by virtue of its casting speed of 1). If, however, the spell could cast with a speed of anything between 0 and 4 the mage might have to think twice before trying that. You wouldn't of course like that situation if you want to be able to predict exact results, but BG is not a game where you can always do that anyway - there is a significant random component to attack rolls, saving throws, damage etc. I actually like the random elements of the game and working out how to manage those successfully in a no-reload context is the major reason why I've played it so much over the years. Personally, I would welcome the challenge if a random initiative factor applied to spell-casting
I think an essential part of the problem is the complexity of the concept of round. It looks like a good understanding of rounds would help a lot to understand this. So, is it possible to know exactly when a round starts and when it finishes? Do we know when the game starts to count rounds and when it stops?
Or maybe a simple practical case might help: Minsc enters a room and sees a Stone Golem. As soon as they see each other they attack each other. However, in order to attack, first they have to run towards each other. When does each round start and finish?
Of course, if this has been discussed elsewhere, just point me to the relevant thread.
Does everyone share the same round?
Why have rounds?
When does a round start?
When do rounds stop?
What activities are not limited by rounds?
Some things are still not clear for me, though: What do you mean by "activities one"?
What do you mean by "activity use"?
What do you mean these are not limited by rounds? The distance you can move in a round is definitely limited. Also the number of times you can attempt stealth, turn undead, etc.
@Bubb ? @kjeron ?
I know the spell doesn't have any Opcode 190 in those spells, but IDK what is packed inside Opcode 16/40.
Thanks!
All these activities are not affected by the general round counter used for what I've termed above as magic-related activities. You're right that some of them have specific time-related counters of their own, though of different forms:
- stealth has a cooldown period such that it can't be tried again for a period after you last dropped out of stealth (or failed to get in it).
- there's no restriction on activating the bard song, shaman dance and turn undead, but the game only checks every 6 seconds whether those are being used so it can take a few seconds before they take effect.
- detect traps and illusions operates in the same way in that the game only actually checks what the ability has done once every 6 seconds. For these skills though there is no text feedback given as to when the check is made. For the bard song etc you are told that and can thus take advantage by doing something else at the same time, e.g. if the bard song is active when you come into contact with an enemy you can wait until the next check is made and then do something else for a few seconds (cast a spell or attack) before re-establishing the song before the next check is made. You can do the same with turn undead.
- the shaman dance is a bit trickier to manipulate as there is a cooldown period after you stop dancing before you can undertake most other activities. One exception to this is that you can still use potions from inside your inventory. Another is that you can still move, though there is an extra step required with this in that any existing spirit summons will disappear if you take any action while dance is activated. You can get round that by turning off the dance as soon as a check is made and only then moving to a new position (and then re-establishing the dance before the next check). With a bit of practice it's not that hard to do that in such a way as to keep the summons engaged with enemies trying to attack you - meaning that shaman spirits are actually a far more useful fighting tactic than generally recognized.
Haste / Slow doesn't appear to affect initiative, but luck does.
That would make sense to me. I assumed that the +/- range would be equal, but just from observation I've always been unsure whether it was +/-2 or +/-3. If the difference is split then perhaps I'm not quite as senile as I thought